Sensory Aroma From Maillard Reaction of Individual and Combination of Amino Acids With Glucase in...

Original article

Sensory aroma from Maillard reaction of individual and

combinations of amino acids with glucose in acidic conditions

Kam Huey Wong, Suraini Abdul Aziz & Suhaila Mohamed*

Faculty of Food Science and Technology, Universiti Putra Malaysia, 43400 Serdang, Malaysia

(Received 20 December 2005; Accepted in revised form 24 August 2006)

Summary The aroma produced in glucose–amino acids (individual and in combination) Maillard reaction, under acidic

conditions at 100 �C were determined and compared by trained panellist. Proline produced pleasant, flowery

and fragrant aroma. Phenylalanine and tyrosine produced dried roses aroma. Alanine produced fruity and

flowery odour, while aspartic acid and serine both produced pleasant, fruity aroma. Arginine, produced a

pleasant, fruity and sour aroma at pH 5.2, but not at its natural pH. Glycine, lysine, threonine and valine

produced a pleasant caramel-like odour. Isoleucine and leucine gave off a burnt caramel aroma. Methionine

developed a fried potato odour. Cysteine and methionine produced savoury, meaty and soy sauce-like

flavours. A combination of these amino acids produced different types of aroma, with the stronger note

dominating the odour of the mixture. This study will help the prediction of flavour characteristics of

hydrolysates from different protein sources.

Keywords: Amino acids, aroma, Maillard reaction, glucose.

Introduction

The parameters that influence the overall flavours andaroma in a Maillard reaction, are the type of amino acidand sugar, pH, temperature, time, moisture content(Lane & Nursten, 1983; Schieberle & Hofmann, 1997)water activity, oxygen, the reaction medium, sulphurdioxide and phosphates (Hurrell & Carpenter, 1977;Namiki, 1988; Shenoy, 1993). Aroma compounds gen-erated from Maillard reaction were mostly studied usingsimple model systems with amino acids (Baltes et al.,1989; Tressl et al., 1989). Many studies have been doneto clarify the mechanisms in organic solvent, rather thanin aqueous solution, and using amines not related to foodingredients instead of amino acids (Hofmann, 1998a).The taste of traditional Japanese foods like miso and

soy sauce is due to the release of amino acids fromnaturally occurring proteins during fermentation. Theproper type and level of free amino acids can signifi-cantly improve the taste of food products in naturallyoccurring or intentionally added flavour potentiators.Although people have been heating sugars and aminoacids at different pHs and temperatures, there is stillno comprehensive study that clearly distinguishes thespecific flavour formation and browning development,especially under very acidic conditions in a Maillardreaction. A reaction between one amino acid and one

sugar will yield hundreds of volatile compounds(Farmer et al., 1989), which include a range of hetero-cyclic compounds having a ring structure containing anatom of N, O or S, depending on the heteroatomspresent in the amino acid, producing more than oneodour in the sensory evaluation.This study attempts to determine and compare the

various aroma resulting from the Maillard reaction ofindividual and combinations of amino acids to relate tothe production of flavours formed in acid-hydrolysedprotein. The sensory results of these amino acids onflavour notes may assist in the prediction of possibleflavour properties of different hydrolysates by referringto the types of amino acids present through amino acidsanalysis.

Materials and methods

Materials

l-alanine, l-arginine hydrochloride, l-aspartic acid,l-cysteine, l-glutamic acid, glycine, l-histidine mono-hydrochloride monohydrate, l-leucine, l-isoleucine,l-lysine monohydrochloride, l-methionine, l-phenylal-anine, l-proline, l-serine, l-threonine, l-tyrosine andl-valine were purchased from Sigma Chemical Co.(Kuala Lumpur, Malaysia). Glucose monohydrate wasobtained from Hamburg Chemical Co. (Hamburg,Germany). Concentrated HCl (37%) and sodium*Correspondent: E-mail: [email protected]

International Journal of Food Science and Technology 2008, 43, 1512–15191512

doi:10.1111/j.1365-2621.2006.01445.x

� 2008 Institute of Food Science and Technology

hydroxide pellets of analytical grade were obtained fromMerck (Kuala Lumpur, Malaysia). Purified water wasused except otherwise stated.

Maillard reaction of amino acids and d-glucose

d-glucose monohydrate (10.0 mmol in 20 mL distilledwater) were reacted with amino acids (3.3 mmol),individually and in combinations in sealed tubes (Hof-mann & Schieberle, 1995) in triplicates, under fivedifferent conditions, i.e.

Sample A: At the original pH of the amino acids and

heated at 100 ± 1 �C in conventional temperature-

controlled thermostatic oven for 24 h;

Sample B: At the original pH of the amino acids and

heated at 100 ± 1 �C for 14 h;

Sample C: At pH 5.2 ± 0.1 and heated at 100 ± 1 �Cfor 24 h;

Sample D: In 6 m HCl as the medium and heated at

100 ± 1 �C for 24 h;

Sample E: In 6 m HCl as the medium and heated at

100 ± 1 �C for 1 h.After the reaction, the tubes were removed and

immediately cooled in ice. Samples D and E were thenneutralized to pH 5.2 ± 0.1. The amounts and types ofamino acids used for the amino acid combinationreaction mixtures are shown in Table 1, with andwithout sulphur amino acids (cysteine and methionine).

These combinations were based on the amino acidprofile of a seed protein acid hydrolysate.Sensory evaluation was carried independently by

eleven trained assessors from the faculty (ASTM,1968, 1981). The descriptors for sensory analysis wereinitially discussed, and a set of reference solutions wasmade. The panel was introduced to these referencesolutions in two training sessions.A set of reference solutions was prepared based on the

odour descriptor set, which consisted of bouillon,meaty, soy sauce, smoky, malty, caramel, beany,chicken, fruity, flowery, pandan, vanilla, prawn, sea-food, chocolate, coffee, buttery and medicinal. Thereference solutions used in order to obtain the mostcharacteristic flavour were:1 Bouillon: one bouillon cube (beef flavour, from Knorr,

CPC/AJI, Kuala Lumpur, Malaysia) dissolved in boiling

water.

2 Meaty: commercial acid hydrolysate (Ajieki, Ajinomoto

(Malaysia) Bhd., Kuala Lumpur, Malaysia) and con-

centrated yeast extract (Vegemite, Kraft Foods Limited,

Victoria, Australia).

3 Soy sauce: soy sauce (Po-Po, Hung Chun Sdn. Bhd.,

Ipoh, Malaysia).

4 Smoky: hickory smoke barbecue sauce (Knorr, CPC/

AJI).

5 Malty: soymilk with malt flavour (Soyfresh, Lam Soon

Singapore Pte Ltd).

Table 1 The types and amounts of amino acids involved in the Maillard reaction of fifteen and seventeen types of amino acids combined

Amino acid

Molecular

weight

Without sulphur amino acids With cysteine and methionine

Concentration of

amino acid used

without sulphur

amino acids (mmol)

Amount of amino acids

used without sulphur

amino acids (g)

Concentration of

amino acid used

with sulphur

amino acids (mmol)

Amount of amino acids

used with sulphur

amino acids (g)

Asp 133.11 0.4599 0.0122 0.3986 0.0106

Glu 147.13 0.6596 0.0194 0.5717 0.0168

Ser 105.09 0.1385 0.0029 0.1200 0.0025

Gly 75.07 0.1970 0.0030 0.1708 0.0026

His 155.18 0.0000 0.0000 0.0000 0.0000

Arg 174.20 0.3116 0.0109 0.2700 0.0094

Thr 119.12 0.1041 0.0025 0.0902 0.0021

Ala 89.09 0.1371 0.0024 0.1188 0.0021

Pro 115.13 0.1011 0.0023 0.0876 0.0020

Tyr 181.19 0.0560 0.0020 0.0486 0.0018

Val 117.15 0.1656 0.0039 0.1435 0.0034

Met 149.20 0 0 0.2200 0.0066

Cys 240.30 0 0 0.2200 0.0106

Ile 131.18 0.1337 0.0035 0.1159 0.0030

Leu 131.18 0.3895 0.0102 0.3376 0.0089

Phe 165.19 0.3002 0.0099 0.2602 0.0086

Lys 146.19 0.1460 0.0043 0.1265 0.0037

Total 3.3000 3.3000

Glucose-amino acids Maillard reaction aromas K. H. Wong et al. 1513

� 2008 Institute of Food Science and Technology International Journal of Food Science and Technology 2008, 43, 1512–1519

6 Caramel: burned sugar.

7 Beany: soymilk (Drinho, Lam Soon Singapore Pte Ltd).

8 Chicken: concentrated chicken stock (Maggi, Nestle

Products Sdn. Bhd., Kuala Lumpur, Malaysia).

9 Fruity: raisin (Big Sister, Madina Setia Sdn. Bhd., Kuala

Lumpur, Malaysia).

10 Flowery: chrysanthemum tea (Yeo’s, Yeo Hiap Seng

(Malaysia) Bhd), Kuala Lumpur, Malaysia.

11 Pandan: pandan flavour essence (Star, Bush Boake Allen

Sdn. Bhd., Kuala Lumpur, Malaysia).

12 Vanilla: vanilla flavour essence (Star, Bush Boake Allen

Sdn. Bhd.).

13 Prawn: prawn flavour 4219 (Matrix Flavours &

Fragrances Sdn. Bhd., Kuala Lumpur, Malaysia).

14 Seafood: seafood flavour powder 1879 (Matrix Flavours

& Fragrances Sdn. Bhd.).

15 Chocolate: chocolate milk (Dutch Lady, Dutch Lady

Milk Ind. Bhd., Kuala Lumpur, Malaysia).

16 Coffee: instant coffee powder (Nescafe Classic, Nestle

Products Sdn. Bhd., Kuala Lumpur, Malaysia) was

diluted in hot water.

17 Buttery: butter (Fern, New Zealand Dairy Board,

Wellington, New Zealand).Colour references were prepared for the samples of

Maillard reaction and assessed visually by the panellistsas follows:1 Dark brown: soy sauce (Po-Po, Hung Chun Sdn. Bhd.).

2 Brown: 20% soy sauce in water.

3 Light brown: 5% soy sauce in water.

4 Very light brown: 2% soy sauce in water.

5 Light yellow: 0.5% soy sauce in water.The references for taste were 1% of the following

ingredients in distilled water:1 Sweet: sucrose (Hamburg).

2 Sour: citric acid (Merck).

3 Salty: sodium chloride (Merck).

4 Bitter: caffeine (Merck).

5 Umami: Monosodium glutamate [Ajinomoto (Malaysia)

Bhd.].The 20-mL samples of the Maillard reaction were

served at room temperature and at 60 �C. All thesamples were coded with 3-digit numbers and served ina randomised order. Six samples were served persession and the experiments were replicated. Thedetection and evaluation of the flavour notes of aminoacids used only simple descriptive test (ISO, 1985). Theresults were collated to produce a list of descriptiveterms applicable to the samples, based on the frequencyof usage of each descriptive word. There were norestrictions as to how many and what terms to use. Theodour descriptions included in the results were onlythose used by four or more panellists (v2 > 5.18;significant at P < 0.05).

Results and discussion

Flavour notes of individual amino acids

Temperature, substrates ratio, pH and time may affectthe flavour notes formed during Maillard reaction.Temperature and substrates ratio were therefore fixed,while pH and time were varied accordingly. Somesamples, such as aspartic acid and glutamic acid had lownatural pH values (pH 3.0 and 3.2 respectively), whilethe others were within the range of pH 5.3–5.8. A varietyof aromas were detected from arginine, aspartic acidand cysteine, implying the suitability of pH 5.2 towardsthe flavour formation. Serine, proline and phenylalaninegave significant fruity and flowery aroma at theiroriginal pH. Alanine produced a flowery odour, whileaspartic acid produced a pleasant odour (Table 2). Thearoma detected from isoleucine, leucine, lysine andthreonine were caramel-like and became more distincton extended heating. No odour was detected fromarginine, glutamic acid and histidine (Table 2), evenafter a colour change. This may be due to the veryminute amounts of volatile odorous compounds re-leased that cannot be detected by the human olfactoryorgan. The detection of odour is dependent on theconcentration and odour threshold of the key odourimpact compounds, and also on the sensitivity of thehuman nose to that particular compound. Although noodour was detected from arginine when Maillard wasperformed at its natural pH values, a pleasant, fruityand sour odour were produced at pH 5.2. Leucine,threonine, tyrosine and valine also released slightlystronger aroma at pH 5.2. No aroma were detected fromglutamic acid and histidine (Tables 2 and 3), probablybecause pH 5.2 were not suitable for the production ofodorous substances from these two amino acids. Undercontrolled pH 5.2 conditions, alanine, serine and aspar-tic acid produced a distinctly pleasant, fruity odour.Cysteine formed a distinct sulphury odour. Glycine,lysine, threonine and valine gave off a pleasant caramel-like odour; while isoleucine and leucine produced adistinct burnt caramel-like aroma. Methionine devel-oped a distinct fried potato odour, while prolineproduced a distinct flowery, pleasant and pandan-likearoma. Phenylalanine and tyrosine gave off a distinctflowery odour almost similar to dried roses. Underuncontrolled pH, with the initial pH 5.2, most aromaformed were almost the same as under pH 5.2 ± 0.1controlled conditions except alanine had additionalpersimmon-like odour, arginine had no odour, glycineand lysine had additional pleasant caramel-like odour,serine had pleasant, slightly ciku, fresh dates-likeodour, proline had additional slightly persimmon-likeodour, phenylalanine had additional dried roses andalmond-like odour, and tyrosine had a slight dried rosesodour.

Glucose-amino acids Maillard reaction aromas K. H. Wong et al.1514

International Journal of Food Science and Technology 2008, 43, 1512–1519 � 2008 Institute of Food Science and Technology

Table 2 Colour and odour of Maillard

products of amino acids and glucose formed

after heating for 14 and/or 24 h at

original pHAmino acid

Original

measured pH

14 and/or 24 hours

Colour Odour

Alanine 5.6 Brown Fruity** (Persimmon), pleasant/sweet**,

flowery**,

Arginine 5.3 Light yellow None, bitter taste*

Aspartic acid 3.0 Very light brown Fruity** (ciku, fresh dates), pleasant/sweet**

Cysteine 4.5 Light yellow Sulphury**

Glutamic acid 3.2 No change None, sour taste**

Glycine 5.7 Brown Caramel-like**, pleasant/sweet*

Histidine 4.0 Light yellow None, slight sweet-sour taste

Isoleucine 5.8 Brown Burnt*, caramel-like*

Leucine 5.7 Brown Burnt**, caramel-like*

Lysine 5.4 Brown Pleasant/sweet*, caramel-like**, bitter taste

Methionine 5.6 Brown Fried potatoes, prawn cracker*

Threonine 5.6 Light brown Pleasant/sweet**, astringent sweet taste

Serine 5.5 Brown Fruity* (slightly ciku, fresh dates),

pleasant/sweet*

Proline 5.7 Light brown Flowery**, pleasant/sweet**, pandan**,

slightly persimmon, bitter taste

Phenylalanine 5.4 Brown Flowery** (dried roses), Almond, bitter taste

Tyrosine 5.6 Brown Flowery** (slightly dried roses), sweet taste

Valine 5.6 Brown Caramel-like**, bitter taste

None (glucose),

control

5.4 No change None

Ciku ¼ Manilkara achras mill.; pandan ¼ Pandanus odorus, Ridl.

*Results from the chi-square distribution showed the flavour, significantly differed from the blank

(P < 0.05, v2 < 4.8), i.e. identified by at least four or five of the eleven panellists.

**Results from the chi-square distribution showed the flavour very significantly differed from the

blank (P < 0.01, v2 < 8.25), i.e. identified by at least six or more of the eleven panellists.



after heating for 24 h at starting pH 5.2

and controlled at pH 5.2 ± 0.1

conditions

Amino acid Colour Odour

Alanine Brown Fruity** (ciku, fresh dates), pleasant/sweet**, flowery**

Arginine Brown Pleasant/sweet**, fruity*, sour*

Aspartic acid Brown Fruity (ciku, fresh dates), pleasant/sweet**, caramel-like**

Cysteine Very light brown Sulphury**, slightly meaty, boiled chickpeas

Glutamic acid Brown None, sour umami taste,

Glycine Brown Pleasant/sweet**, flowery*

Histidine Light yellow None, sour taste

Isoleucine Brown Burnt**, caramel-like**

Leucine Brown Burnt**, caramel-like*, biscuit-like

Lysine Brown Pleasant/sweet**, pandan*,bitter taste

Methionine Brown Fried potatoes, prawn crackers*

Threonine Light brown Pleasant/sweet**, fruity**

Serine Brown Pleasant/sweet**

Proline Brown Flowery*, pleasant/sweet*, pandan**

Phenylalanine Brown (cloudy) Flowery** (roses), almond, Mimusops elengi flower**

Tyrosine Brown Flowery*, fruity*, pleasant/sweet, tea-like

Valine Brown Caramel-like**, biscuit-like, malty, chocolate, bitter taste

None (glucose) No change None

Ciku ¼ Manilkara achras mill.; pandan ¼ Pandanus odorus, Ridl.



**Results from the chi-square distribution showed the flavour very significantly differed from the




During the Strecker degradation, sulphur-containingamino acids will lose one carbon as CO2 to form highlyodorous thioaldehydes. For example, methionine willform methional, which produces a significant potatoflavour (Self, 1967) as similarly found for methionine inthis study (Tables 2–4). Compounds of particularimportance for the development of ‘meaty’ flavours,reportedly have a furan or thiophene ring with a thiolgroup in the 3-position, while similar compounds with athiol in the 2-position tend to be ‘burnt’ and ‘sulphur-ous’ (Farmer, 1994). In this study, cysteine was found toemit a sulphury odour at its natural pH, and bothsulphury and meaty aroma in acidic pH.Maillard-type reaction between the amino acid pro-

line and reducing carbohydrates is well known forgenerating popcorn-like, roasty aroma upon thermaltreatment (Hunter et al., 1969). Nine key odourants aregenerated in thermally treated proline/glucose reactionmixtures, of which one gave buttery odour, another gavecaramel-like odour, six gave popcorn-like odour and thelast unknown odourant gave a roasty odour (Hofmann& Schieberle, 1998a). The distillate obtained by boilingand simultaneously extracting an aqueous mixture ofproline and d-glucose, elicited an intense roasty, pop-corn-like odour (Hofmann & Schieberle, 1998a). Thedifference in the flavour developed from proline-glucoseinteraction between this study and those reportedpreviously is probably due to the acid pH of the

reaction medium, and that the panellists were morefamiliar with pandan than popcorn flavour.The factors influencing the generation of aroma from

cysteine/carbohydrate reaction mixtures are still notfully understood, particularly because the contributionof a single odorant to the key flavour notes has not beenestablished. A sensory evaluation of cysteine/glucosemixture revealed meat-like and pungent odour notes asthe predominant odour qualities, while a cysteine/rhamnose mixture revealed roasty, meat-like andsulphury odour notes with the highest intensities andcysteine/ribose mixture produced an additional caramel-like and a seasoning-like odour note (Salter et al., 1988;Hofmann & Schieberle, 1998b). Similar results werefound in this study.

Effect of extreme low pH

At 6 m HCl (which is the concentration of HCl used inproducing acid protein hydrolysates), most samplesturned brown after 1 h at 100 �C (Table 4). Cysteinewas dark brown in colour after the 1-h heating, andgradually turned black after 24 h. Tyrosine also exhib-ited colour change from light to a darker colour withprolonged heating. Results clearly show that 1-h heatingat an extremely low pH is sufficient, and some samplesproduced a burnt odour. Other samples such as aspar-tic acid, cysteine, methionine, threonine, proline and



after heating for 1 h in the presence of 6 m

HCl (pH < 0.1)

Amino acid Colour Odour

Alanine Brown Pleasant/sweet**, fruity* (ciku, fresh dates), caramel-like**,

slightly burnt**

Arginine Brown Caramel-like*, burnt**, slight sweet/pleasant*

Aspartic Acid Brown Pleasant/sweet**, fruity (ciku, fresh dates), caramel-like**

Cysteine Dark brown Sulphury**, meaty*

Glutamic Acid Brown Pleasant/sweet**, caramel-like**, burnt**

Glycine Brown Pleasant/sweet**, caramel-like**, burnt**

Histidine Brown Caramel-like**, burnt*

Isoleucine Brown Burnt**, coffee-like*, prune

Leucine Brown Burnt*, coffee-like*, caramel-like*, malty*

Lysine Brown Pleasant/sweet**, cardboard, Herbal tea*

Methionine Brown Meaty**, sulphury**, fried potatoes*

Threonine Brown Pleasant/sweet**, fruity** (ciku, fresh dates), flowery*,

caramel-like**

Serine Brown Slightly burnt**, pleasant/sweet*

Proline Brown Flowery**, pleasant/sweet**, pandan*, slightly alkaline

Phenylalanine Brown Flowery** (roses)

Tyrosine Brown Pleasant/sweet*, caramel-like*, burnt**

Valine Brown Pleasant/sweet**, caramel-like**, burnt*, malty

Ciku ¼ Manilkara achras mill.; pandan ¼ Pandanus odorus, Ridl. All products contained black

sediment. After 24 h most of the samples had burnt, smoky aroma, except phenylalanine and

proline samples retained their pleasant, flowery aroma.



**Results from the chi-square distribution showed the flavour, very significantly differed from the




phenylalanine emitted a pleasant odour. When theheating period was prolonged to 24 h, more sampleswith the exception of methionine, proline and phenyl-alanine emitted burnt odour.

Serving temperature

Table 5 shows that while serving temperature had noeffect on the colour, apparent differences in odourwere detected. Glutamic acid and histidine had nodetectable odour when served at room temperature,but when served at 60 �C, odour was easily detected.Glycine, lysine and serine also emitted additionalaroma. This might be due to the minute amounts ofvolatiles probably hydrogen bonded to the aqueousmedium which made detection at room temperaturedifficult.

Flavour notes of amino acids combinations

Amino acids combinations heated at their natural pHvalues at 100 �C for 24 h produced an odour similar tosoy sauce, hydrolysed yeast extract (Marmite) andslightly of dried salted fish (Table 6). When they areheated in the presence of 6 m HCl as for the productionof acid-hydrolysed proteins, a strong chemical-likedisinfectant and chlorine-like aroma were produced.This may be due to the usage of a strong acid in thereaction. Heating at pH 5.2, produced a burnt, smokedand slightly soy sauce-like aroma. Cysteine and methi-onine may be involved in the production of soy sauceodour, which may have been destroyed when heated inthe presence of 6 m HCl. A good odour was producedunder a very strong acidic condition, after just 1 h ofheating at 100 �C. Heating the amino acids at

Table 5 Comparison of colour and odour of Maillard products of amino acids and glucose heated and maintained at pH 5.2 ± 0.1 served at

room temperature and 60 �C

Served at room temperature Served at 60 �C

Amino acid Colour Odour Colour Odour

Arginine Brown Pleasant/sweet**, fruity*, sour* Brown Pleasant/sweet**, caramel-like**, fruity, sour

Glutamic acid Brown None**, fruity sour umami taste Brown Pleasant/sweet**, caramel-like*, biscuit-like*

Glycine Brown Pleasant/sweet**, flowery* Brown Pleasant/sweet**, flowery*, caramel-like**

Histidine Light yellow None**, sour taste Light yellow Pleasant/sweet**, caramel-like*

Lysine Brown Pleasant/sweet**, pandan* Brown Pleasant/sweet**, pandan*, flowery*

Threonine Very light brown Pleasant/sweet**, fruity** Very light brown Pleasant/sweet**, fruity*

Serine Brown Pleasant/sweet** Brown Pleasant/sweet**, caramel-like**

Tyrosine Brown Flowery*, fruity*, Pleasant/sweet*,

tea-like

Brown Flowery** (roses), fruity, Pleasant/sweet, tea-like

Valine Brown Caramel-like**, biscuit-like, malty,

chocolate

Brown Caramel-like**, biscuit-like, malty,

chocolate, burnt**

*Results from the chi-square distribution showed the flavour, significantly differed from the blank (P < 0.05, v2 < 4.8), i.e. identified by at least four or

five of the eleven panellists.

**Results from the chi-square distribution showed the flavour, very significantly differed from the blank (P < 0.01, v2 < 8.25), i.e. identified by at least six

or more of the eleven panellists.

Table 6 Mixture of aroma detected from the Maillard reaction of a combination of amino acids (as in Table 1 in the presence of sulphur

amino acids) with glucose under five different reaction conditions

Samples Description of odour

A: At the original pH of the amino acids and heated at 100 ± 1 �C in

conventional temperature-controlled thermostatic oven for 24 h

Soy sauce**, marmite*, salted fish-like*

B: At the original pH of the amino acids and heated at 100 ± 1 �C for 14 h Chemical*, strong disinfectant*, chlorine-like*

C: At pH 5.2 ± 0.1 and heated at 100 ± 1 �C for 24 h Burnt*, smoky*, slightly soy sauce-like*

D: In 6 M HCl as the medium and heated at 100 ± 1 �C for 24 h.

Neutralised to pH 5.2 ± 0.1 after immediate cooling in ice

Flowery*, caramel-like*

E: In 6 M HCl as the medium and heated at 100 ± 1 �C for 1 h.

Neutralised to pH 5.2 ± 0.1 after immediate cooling in ice

Chemical*, strong disinfectant*, chlorine-like*

*Results from the chi-square distribution showed the flavour, significantly differed from the blank (P < 0.05, v2 < 4.8), i.e. identified by at least four or

more of the eleven panellists.

**Results from the chi-square distribution showed the flavour, very significantly differed from the blank (P < 0.01, v2 < 8.25), i.e. identified by at least six

or more of the eleven panellists.



100 ± 1 �C for 14 h, under the original pH, released aflowery and caramel-like aroma. No soy sauce-likeflavours were released when cysteine and methioninewere absent. The sensation of odour is produced by thevolatile chemical substances, which stimulate the recep-tors in the nasal epithelium.A combination of these amino acids produced differ-

ent types of aroma. The types of odour produced were acombination of aroma of each individual amino acidwith the stronger note dominating and becoming themain odour of the samples (Table 6). For example,sample D produced a flowery odour, most probablycaused by phenylalanine and tyrosine. When cysteineand methionine were added to the reaction composition,the main odour was different. It produced soy sauce,hydrolysed yeast extract (Marmite) and slightly saltedfish-like aroma as in sample A of Table 6. On the otherhand, the lack of cysteine and methionine in samples Dand E caused the production of a non-soy sauce odour.This clearly shows that these two sulphur-containingamino acids (cysteine and methionine) are the mainamino acids that are responsible for the production of ameaty flavour. The determined flavour notes of theamino acids Maillard reactions are closely related to theflavour produced by vegetable protein hydrolysates.

Colour

Glutamic acid did not exhibit any colour changes at itsoriginal pH even after heating for 24 h (Table 2). TheMaillard reaction may have already taken place longbefore any colour changes were observed. Under con-trolled pH, the mixture turned brown in colour at theend of the reaction. Variation of odour was moreprominent than the colour changes under both reactionconditions. Most samples had similar colours whenreacted at different pH values for 24 h, as this longperiod of time is believed to allow maximum interactionof the substrates. Arginine, aspartic acid, cysteine,glutamic acid and proline exhibited prominent colourchanges especially at pH 5.2.The complexity of the nonenzymatic browning reac-

tions is known to be at least partly due to the sugarcaramelisation processes (Greenshields, 1973). In thisstudy, glucose solution without amino acid was alsotreated under similar conditions to examine the appear-ance of the brown colour. Sensory results show nochanges in colour when glucose was present alone in thereaction medium at approximately pH 5.4, as previouslyreported (Ajandouz & Puigserver, 1999), and that theintensity of browning increased with pH values, and thepresence of almost all of the essential amino acids. Noneof these amino acids, with the exception of tryptophan,gave rise to any browning in the absence of glucose.Maillard reaction is temperature dependent and the

reaction rate may increase two to three times for each

10 �C rise in temperature of model systems (Birch,1977), and even more than this in the natural systems.Up to 60 �C, browning is normally a zero orderreaction, but at higher temperatures, changes mayfollow a first order reaction (Labuza & Saltmarch,1981). The effect of temperature on Maillard reaction, isalso related to other variables such as acidity and wateractivity. Activation energy decreases with increasingwater activity, resulting in increasing browning rate.Reaction at 37 �C occurred over a period of days, whileat above 100 �C, reaction time occurred within minutes.In a mixture of albumin and glucose, 76% of the e-amino lysine were unavailable after 30 days at 37 �C,compared with 85% in 15 min at 121 �C (Armstrong,1994; Mustapha, 1997). pH value has a major influenceon many important pathways of the Maillard reactionand the products (Ames, 1988). The browning rate ofglucose/glycine peptides below pH 6 is greater thanglucose/glycine at the same molar concentration, but thereverse was observed at a higher pH (Labuza & Schmidl,1986). Increasing the pH values of the reaction mediumwill generally enhance the reaction (Labuza et al., 1983;Ashoor & Zent, 1984; Petriella et al., 1985). The typeand the amount of final products may differ if theexperimental conditions are not exactly identical (Ameset al., 1997), and if the pH is not controlled during thereaction (Ames et al., 1993). The present results agreewith the previous researchers’ observations, and newaromas are reported here due to the slight differences inreaction conditions.The likelihood of mutagenic or carcinogenic com-

pounds formed under the conditions of the reaction isbeyond the scope of this study. The formation andoccurrence of carcinogenic heterocyclic amines in glu-cose–amino acids model systems have been reported(Johansson et al., 1995) and reviewed (Skog et al.,1998).

Conclusions

In this study, most amino acids produced a pleasant oracceptable flavour with only a few producing unpleasantburnt and sulphury aroma. Consequently, the Maillardreaction can be used as a basis for the production ofspecific flavouring products by carefully selecting thesugars and amino acids, and controlling the processingconditions within narrow specifications. Both pH andtime play an important role in controlling the type ofodorous compounds that are produced and the use ofextreme pH, such as 6 m HCl would produce a burntodour even with only an hour of heating. Prolongedheating time under this condition would cause theunpleasant odour to become even more intense. Com-bination of amino acids produced different results.Slightly acidic amino acid mixtures heated for 24 h(Table 6, sample C) produced a better odour than amino



acid mixtures heated at their natural pH for (sample B)14 h. The serving temperature also plays an importantrole in the detection of odour. At 60 �C, higherconcentrations of volatile compounds are being released.

References

Ajandouz, E.H. & Puigserver, A. (1999). Nonenzymatic browningreaction of essential amino acids: effect of pH on caramelization andMaillard reaction kinetics. Journal of Agricultural and FoodChemistry, 47, 1786–1793.

Ames, J.M. (1988). The Maillard browning reaction – an update.Chemistry and Industry, 5, 558–561.

Ames, J.M., Apriyantono, A. & Arnoldi, A. (1993). A low molecularweight coloured compounds formed in xylose-lysine model system.Food Chemistry, 46, 121–127.

Ames, J.M., Arnoldi, A., Bates, L. & Negroni, M. (1997). Analysis ofthe methanol-extractable of a model extrusion-cooked cerealproduct. Journal of Agricultural and Food Chemistry, 45, 1256–1263.

Armstrong, H.J. (1994). Enhanced protein gelation using the Maillardreaction and elevated temperatures. PhD Thesis, University ofNottingham, UK.

Ashoor, S.H. & Zent, J.B. (1984). Maillard browning in commonamino acids and sugars. Journal of Food Science, 49, 1206–1207.

ASTM (1968). ASTM Committee E-18 on Sensory Evaluation ofMaterials and Products. ASTM STP 434. In: J.B. Wheeler, H.M.Hoersch, H.P. Mahy & A.S. Kleinberg (eds), Manual on SensoryTesting Methods. Pp. 7–8. Philadelphia, PA: American Society forTesting and Materials.

ASTM (1981). ASTM Committee E-18 on Sensory Evaluation ofMaterials and Products. ASTM STP 758. In: Guidelines for theSelection and Training of Sensory Panel Members (edited by J.B.Wheeler, H.M. Hoersch, H.P. Mahy & A.S. Kleinberg). Pp. 5–28.Philadelphia, PA: American Society for Testing and Materials.

Baltes, W., Kunert-Kirchhoff, J. & Reese, G. (1989). Model reactionson generation of thermal aroma compounds. In: Thermal Generationof Aromas (edited by T.H. Parliment, R.J. McGorrin & C.T. Ho).Pp. 141–155. ACS Symposium Series 409. Washington, DC:American Chemical Society.

Birch, G.G. (1977). Chemical, physical and biological changes incarbohydrates induced by thermal processing. In: Physical, Chemicaland Biological Changes in Food Caused by Thermal Processing(edited by T. Hoyem & O. Kvale). Pp. 152–167. London: AppliedScience Publishers Limited.

Farmer, L.J. (1994). The role of nutrients in meat flavour. Proceedingsof the Nutrition Society, 53, 327–333.

Farmer, L.J., Mottram, D.S. & Whitfield, F.B. (1989). Volatilecompounds produced in Maillard reactions involving cysteine,ribose and phospholipid. Journal of the Science of Food andAgriculture, 49, 347–368.

Greenshields, R.N. (1973). Caramel-part 2. Manufacture, compositionand properties. Process Biochemistry, 4, 17–20.

Hofmann, T. (1998a). Characterization of the chemical structure ofnovel colored Maillard reaction products from furan-2-aldehydeand amino acids. Journal of Agricultural and Food Chemistry, 46,932–940.

Hofmann, T. & Schieberle, P. (1995). Evaluation of the key odorantsin a thermally treated solution of ribose and cysteine by aromaextraction dilution techniques. Journal of Agriculture and FoodChemistry, 43, 2187–2194.

Hofmann, T. & Schieberle, P. (1998a). Flavor contribution andformation of the intense roast-smelling odorants 2-propionyl-1-pyrroline and 2-propionyltetrahydropyridine in Maillard-type reac-tions. Journal of Agriculture and Food Chemistry, 46, 2721–2726.

Hofmann, T. & Schieberle, P. (1998b). Identification of key aromacompounds generated from cysteine and carbohydrates under

roasting conditions. Zeitschrift fur Lebensmittel Untersuchung und-Forschung A, 207, 229–236.

Hunter, I.R., Walden, M.K., Scherer, J.R. & Lundin, R.E. (1969).Preparation and properties of 1,4,5,6-tetrahydro-2-acetopyridine, acracker-odor constituent of bread aroma. Cereal Chemistry, 46, 189–195.

Hurrell, R.F. & Carpenter, K.J. (1977). Maillard reactions in foods. In:Physical, Chemical and Biological Changes in Food Caused byThermal Processing (edited by T. Hoyem & O. Kvale). Pp. 168–184.London: Applied Science Publishers Limited.

ISO (1985). Sensory Analysis – Methodology – General Guidance, 1stedn. International Standard 6658, Ref. No. ISO 6658–1985 (E). Pp.8–9. Switzerland, Geneve: International Organization for Standard-ization.

Johansson M.A.E., Fay L.B., Gross G.A., Olsson K. & Jagerstad M.(1995). Influence of amino acids on the formation of mutagenic/carcinogenic heterocyclic amines in a model system. Carcinogenesis,16, 2553–2560.

Labuza, T.P. & Saltmarch, M. (1981). The nonenzymatic browningreaction as affected by water in foods. In: Water Activity: Influenceson Food Quality (edited by L.B. Rockland & G.F. Stewart). Pp. 605–650. New York: Academic Press.

Labuza, T.P. & Schmidl, M.K. (1986). Advances in the control ofbrowning reactions in foods. In: Role of Chemicals in the Quality ofProcessed Food (edited by O.R. Fennema, W.H. Chang & C.Y. Lii).Pp. 65–95. Westport, CT: Food and Nutrition Press.

Labuza, T.P., Loncin, M. & Weiss, M. (1983). Investigations into theinfluence of water activity, pH and heat treatment on the velocity ofthe Maillard reaction in foods. Journal of Food Quality, 6, 103–118.

Lane, M.C. & Nursten, H. E. (1983). The variety of odors produced inMaillard model systems and how they are influenced by the reactionconditions. In: G.R. Waller & M.S. Feather (eds), The MaillardReaction in Food and Nutrition. ACS Symposium Series 215. Pp 141–158. Washington, DC: American Chemical Society.

Mustapha, W.A.W. (1997). Influence of the physico-chemical envi-ronment as product formation in the Maillard reaction. PhD Thesis,University of Nottingham, UK.

Namiki, M. (1988). Chemistry of Maillard reactions: recent studies onthe browning reaction mechanism and the development of antioxi-dants and mutagens. Advances in Food Research, 32, 115–143.

Petriella, C., Resnik, S.L., Lozano, R.D. & Chirife, J. (1985). Kineticsof deteriorative reactions in model food systems of high wateractivity: color changes due to non enzymatic browning. Journal ofFood Science, 50, 622–625.

Salter, L.J., Mottram, D.S. & Whitfield, F.B. (1988). Volatilecompounds produced in Maillard reactions involving glycine, riboseand phospholipid. Journal of the Science of Food and Agriculture, 46,227–242.

Schieberle, P. & Hofmann, T. (1997). Identification of the characterimpact odorants in fresh strawberry juice by quantitative measure-ments and sensory studies on model mixtures. Journal of Agricultureand Food Chemistry, 45, 227–232.

Self, R. (1967). Potato flavor. In: Chemistry and Physiology of Flavors(edited by H.W. Schultz, E.A. Day & L.M. Libbey). Pp 362–389.Westport, CT: AVI Publishing Company, Inc.

Shenoy, V.R. (1993). Maillard reactions and its applications. Journalof Scientific and Industrial Research, 52, 684–689.

Skog K.I., Johansson M.A.E. & Jagerstad M.I. (1998). Carcinogenicheterocyclic amines in model systems and cooked foods: a review onformation, occurrence and intake. Food and Chemical Toxicology,36, 879–896.

Tressl, R., Helak, B., Martin, N. & Kersten, E. (1989). Formation ofamino acid specific Maillard products and their contribution inthermally generated aromas. In: Thermal Generation of Aromas(edited by T.H. Parliment, R.J. McGorrin & C.T. Ho). ACSSymposium Series 409. Pp. 156–171. Washington, DC: AmericanChemical Society.



Sensory Aroma From Maillard Reaction of Individual and Combination of Amino Acids With Glucase in...

Documents

Transcript of Sensory Aroma From Maillard Reaction of Individual and Combination of Amino Acids With Glucase in...